Lithium secondary batteries, which have recently been utilized as main power supplies of mobile communication devices and potable electric appliances, are characterized by having high electromotive force and a high energy density.
Lithium secondary batteries in which lithium metal is used as a negative electrode material has a high energy density, but dendrite deposits on a negative electrode during charging. As charge/discharge are repeated, the dendrite develops to penetrate a separator and comes into contact with a positive electrode, causing an internal short circuit. Further, as having a large surface area, the deposited dendrite has a high reactivity and reacts, on the surface thereof, with a solvent in an electrolyte to form a solid-electrolytic interface layer having poor electronic conductivity. This leads to higher internal resistance within the battery and to the existence of particles isolated from the electronically conductive network. For these reasons, a charge/discharge efficiency decreases.
Lithium secondary batteries in which lithium metal is used as a negative electrode material therefore have problems of low reliability and short cycle life.
As an alternative negative electrode material for lithium metal currently used has been a carbon material capable of absorbing and desorbing lithium ions, and batteries using this for the negative electrode thereof have come into practical use. There normally occurs no problem of an internal short circuit due to dendrite in a negative electrode using a carbon material since metallic lithium does not deposit thereon. However, the theoretical capacity of graphite, one of carbon materials, is 372 mAh/g, which is as small as about one tenth of the theoretical capacity of Li metal.
As other negative electrode materials known is a elemental metallic or non-metallic material, which form a compound with lithium. In the case of silicon, tin and zinc, for example, the compositions of the respective compounds with the largest content of lithium are Li22Si5, Li22Sn5 and LiZn. With lithium contained in this degree in each of the compounds, metallic lithium usually does not deposit, to cause no problem of the internal short circuit due to dendrite. Electrochemical capacities obtained by the change from elemental substances to these compounds are, respectively, 4199 mAh/g, 993 mAh/g and 410 mAh/g, which are all larger than the theoretical capacity of graphite.
As a negative electrode material using a different compound from those described above, Japanese Laid-Open Patent Publication No. Hei 7-240201 proposes a silicate of non-iron metal including a transition element. Further, Japanese Laid-Open Patent Publication No. Hei 9-63651 proposes a compound which comprises an intermetallic compound containing the Group 4B elements, and at least one of P and Sb, and whose crystal structure is any of CaF2 type, ZnS type and AlLiSi type.
However, the aforesaid negative electrode materials with higher capacities than those of carbon materials have problems as described below.
Simple substance metals and simple substance non-metals, which form a compound with lithium, have inferior charge/discharge cycle characteristics to carbon materials. The reason for that is presumed as follows.
For example, silicon contains 8 silicon atoms in the crystallographic unit lattice thereof (cubic, space group Fd-3 m). From a lattice constant “a”=0.5420 nm determined are a unit lattice volume of 0.1592 nm3 and a volume per one silicon atom of 19.9×10−3 nm3. Judging from the phase diagram of the silicon-lithium binary type, it is considered that, when silicon electrochemically reacts with lithium at room temperature to form a compound, two phases, silicon and the compound Li12Si7, coexist in the initial stage of the reaction. Li12Si7 contains 56 silicon atoms in the crystallographic unit lattice thereof (rhombic, space group Pnma). From lattice constants “a”=0.8610 nm, “b”=1.9737 nm and “c”=1.4341 nm, a unit lattice volume of 2.4372 nm3 and a volume per one silicon atom (a value obtained by dividing the unit lattice volume by the number of the silicon atoms in the unit lattice) of 43.5×10−3 nm3 are determined. Hence conversion from silicon to the compound Li12Si7 causes an increase in volume by 2.19 times and the material thus expands. When the reaction proceeds in such a state as the two phases, silicon and the compound Li12Si7, coexist, silicon partly converts into the compound Li12Si7 and a large volume difference therebetween causes serious distortion of the material. This material is therefore considered as prone to cracking and pulverizing.
Moreover, as the electrochemical reaction between silicon and lithium advances, the compound Li22Si5 with the largest content of lithium is ultimately obtained. Li22Si5 contains 80 silicon atoms in the crystallographic unit lattice thereof (cubic, space group F23). From a lattice constant “a”=1.875 nm determined are a unit lattice volume of 6.5918 nm3 and a volume per one silicon atom (a value obtained by dividing the unit lattice volume by the number of the silicon atoms in the unit lattice) of 82.4×10−3 nm3. Hence conversion from silicon to the compound Li22Si5 causes an increase in volume by 4.14 times, and hence the material expands significantly. In the discharge reaction of the negative electrode material, on the other hand, lithium is gradually reduced from the compound, and the material thus shrinks. It is therefore considered that a major variation in material volume during charging/discharging brings about significant distortion of the material, whereby cracking occurs to pulverize particles.
It is further considered that, as spaces are formed among the pulverized particles to cause segmentation of the electronically conductive network, a portion incapable of being involved in the electrochemical reaction increases to deteriorate the charge/discharge characteristic.
Tin contains 4 tin atoms in the crystallographic unit lattice thereof (tetragonal, space group I41/amd). From lattice constants “a”=0.5820 nm and “c”=0.3175 nm, a unit lattice volume of 0.1075 nm3 and a volume per one tin atom of 26.9×10−3 nm3 are determined. Judging from the phase diagram of the tin-lithium binary type, it is considered that, when tin electrochemically reacts with lithium at room temperature to form a compound, two phases of tin and the compound Li2Sn5 coexist in the initial stage of the reaction. Li2Sn5 contains 10 tin atoms in the crystallographic unit lattice thereof (tetragonal, space group P4/mbm). From lattice constants “a”=1.0274 nm and “c”=0.3125 nm, a unit lattice volume of 0.32986 nm3 and a volume per one tin atom (a value obtained by dividing the unit lattice volume by the number of the tin atoms in the unit lattice) of 33.0×10−3 nm3 are determined. Hence conversion from tin to the compound Li2Sn5 causes an increase in volume by 1.23 times and the material thus expands.
Moreover, as the electrochemical reaction between tin and lithium advances, the compound Li22Sn5 with the largest content of lithium is ultimately obtained. Li22Sn5 contains 80 tin atoms in the crystallographic unit lattice thereof (cubic, space group F23). From a lattice constant “a”=1.978 nm determined are a unit lattice volume of 7.739 nm3 and a volume per one tin atom (a value obtained by dividing the unit lattice volume by the number of the tin atoms in the unit lattice) of 96.7×10−3 nm3. Hence conversion from tin to the compound Li22Si5 causes an increase in volume by 3.59 times, and hence the material expands significantly.
Zinc contains 2 zinc atoms in the crystallographic unit lattice thereof (hexagonal, space group P63/mmc). From lattice constants “a”=0.2665 nm and “c”=0.4947 nm, a unit lattice volume of 0.030428 nm3 and a volume per one zinc atom of 15.2×10−3 nm3 are determined. Judging from the phase diagram of the zinc-lithium binary type, when zinc electrochemically reacts with lithium at room temperature to form several compounds, the compound LiZn with the largest content of lithium is ultimately obtained. LiZn contains 8 zinc atoms in the crystallographic unit lattice thereof (cubic, space group Fd-3 m). From lattice constants “a”=0.6209 nm determined are a unit lattice volume of 0.2394 nm3 and a volume per one zinc atom (a value obtained by dividing the unit lattice volume by the number of the zinc atoms in the unit lattice) of 29.9×10−3 nm3. Hence conversion from zinc to the compound LiZn causes an increase in volume by 1.97 times and the material thus expands.
In the case of using tin or zinc, as in the case of silicon, therefore, the volume variation in negative electrode material due to the charge/discharge reaction is large and the variation continues in a state where the two phases with great volume differences coexist. This is considered as the cause of cracking of a material and pulverization of the particles thereof. It is further thought that, as spaces are formed among the pulverized particles to cause segmentation of the electronically conductive network, a portion incapable of being involved in the electrochemical reaction increases to deteriorate the charge/discharge characteristic.
That is to say, when a simple substance metal or a simple substance non-metal, which forms a compound with lithium, is used for a negative electrode, the metal or non-metal suffers a large volume variation and tends to be pulverized. This presumably causes the inferior charge/discharge cycle characteristic to a negative electrode using a carbon material.
Other than the aforesaid simple substances, Japanese Laid-Open Patent Publication No. Hei 7-240201 proposes a silicate of non-iron metal including a transition element as a negative electrode material capable of improving the cycle life characteristic. In this publication provided are examples of batteries in which a silicate of non-iron metal including a transition element is used as a negative electrode material and comparative examples of batteries in which lithium metal is used as a negative electrode material, and the charge/discharge cycle characteristics of the respective batteries are compared. It is then disclosed that the charge/discharge characteristics of the batteries in the examples are improved more than those of the batteries in the comparative examples. By comparison with a battery in which natural graphite is used as a negative electrode material, however, the maximum increase in battery capacity in the examples is only about 12%.
Although not definitely stated in the publication, therefore, there appears to be no significant increase in capacity of the battery in which silicate of non-iron metal including a transition metal for the negative electrode thereof is used, as compared with a battery in which graphite is used for the negative electrode thereof.
Further, Japanese Laid-Open Patent Publication No. Hei 9-63651 proposes a compound which comprises an intermetallic compound containing the Group 4B element and at least one of P and Sb, and whose crystal structure is any of CaF2 type, ZnS type and AlLiSi type, as a negative electrode material capable of improving the cycle life characteristic.
It is disclosed that the charge/discharge cycle characteristic is more improved in an example where the aforesaid compound is used for the negative electrode than in a comparative example where Li—Pb alloy is used for the negative electrode. It is further disclosed that a higher capacity is obtained in the example than in a case of using graphite for the negative electrode.
However, the battery in the example exhibits a significant decrease in discharge capacity at 10th to 20th cycles, and even in the case of using Mg2Sn, which is presumably the most favorable compound, the discharge capacity decreases to about 70% of the initial capacity after about 20th cycles.
Furthermore, Japanese Laid-Open Patent Publication No. 2000-30703 proposes a negative electrode material, comprising solid phases A and B, the solid phase A comprising at least one of silicon, tin and zinc as the constituent element thereof, the solid phase B comprising a solid solution or an intermetallic compound, containing one of silicon, tin and zinc as the constituent element of the solid phase A, and at least one element selected from the group consisting of the elements of Group 2, transition, Group 12, Group 13, and Group 14 which are listed in Long Form of Periodic Table, with carbon excluded from Group 14 element. It is disclosed that a battery using this negative electrode material for the negative electrode thereof has a higher capacity and a more improved cycle life characteristic than a battery using graphite for the negative electrode thereof.
When the crystallinity of the solid phase A in this material is high, a problem may arise that stress within particles at the time of absorbing lithium concentrates in one direction to make the particles prone to destruction, leading to a shorter cycle life.
A description is given to crystallinity. In general, the crystal properties are largely classified into amorphousness (in a state where a diffraction line is not obtained in a wide-angle X-ray diffraction measurement), micro-crystalline, poly-crystalline and mono-crystalline. For solving the aforesaid problems required is to lower the crystallinity of the solid phase A. That the solid phase A in a low-crystallinity state here means that the solid phase A is in a mixed state of amorphousness and micro-crystalline. It is to be noted that micro-crystalline means poly-crystalline with a crystal size of not larger than about 150 nm. Further, the crystallinity of the solid phase B may be poly-crystalline or micro-crystalline.
In order to solve the aforesaid problems, an object of the present invention is to provide a negative electrode material capable of suppressing pulverization thereof due to repeated cycles. Another object of the present invention is to provide a non-aqueous electrolyte secondary battery having a high capacity and an excellent cycle life characteristic, by the use of this negative electrode material.